For example, photographed images of each inner organ are outputted as gray scale brightness value information from a medical image apparatus such as a Computed Tomography (CT) apparatus, Magnetic Resonance Imaging (MRI) apparatus, or ultrasonic echo apparatus. As a method for rendering volume data that is three-dimensional data, which is obtained by accumulating such images in a perpendicular direction of the images at regular intervals or obtained by simulation using a structured mesh, a volume rendering is known.
Moreover, as a method similar to the volume rendering, a texture-based volume rendering is also known. In this texture-based volume rendering, plural cross sections for the volume data are set to display the volume data as a pattern by interpolating adjacent values. In this method, because the display accuracy depends on the number of cross-sections, it is possible to reduce the processing loads if the number of cross-sections is lesser. However, in case where the number of cross-sections is lesser, when the volume data is rotated in an arbitrary direction in order to change the sight line, the display accuracy may be worsened because the step caused by the large difference between the contents of the cross sections is emphasized in the image.